Committment-Based Data-Aware Multi-Agent-Contexts Systems
Communication and interaction among agents have been the subject of extensive investigation since many years. Commitment-based communication, where communicating agents are seen as a debtor agent who is committed to a creditor agent to bring about something (possibly under some conditions) is now very well-established. The approach of DACMAS (Data-Aware Commitment-based MAS) lifts commitment-related approaches proposed in the literature from a propositional to a first-order setting via the adoption the DRL-Lite Description Logic. Notably, DACMASs provide, beyond commitments, simple forms of inter-agent event-based communication. Yet, the aspect is missing of making a MAS able to acquire knowledge from contexts which are not agents and which are external to the MAS. This topic is coped with in Managed MCSs (Managed Multi-Context Systems), where however exchanges are among knowledge bases and not agents. In this paper, we propose the new approach of DACmMCMASs (Data-Aware Commitment-based managed Multi- Context MAS), so as to obtain a commitment-based first-order agent system which is able to interact with heterogeneous external information sources. We show that DACmMCMASs retain the nice formal properties of the original approaches.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
ContextAgent: Context-Aware Proactive LLM Agents with Open-World Sensory Perceptions
Recent advances in Large Language Models (LLMs) have propelled intelligent agents from reactive responses to proactive support. While promising, existing proactive agents either rely exclusively on observations from encl…
Context-Aware RL for Agentic and Multimodal LLMs
Large language models (LLMs) often fail when answering requires identifying a small but decisive piece of evidence within a long or complex context, such as a single line in a tool trace or a subtle detail in an image. W…
Visual Question AnsweringReinforcement LearningMultimodal ReasoningProAgent: Harnessing On-Demand Sensory Contexts for Proactive LLM Agent Systems in the Wild
Recent studies have begun to explore proactive large language model (LLM) agents that provide unobtrusive assistance by automatically leveraging contextual information, such as in code editing and in-app suggestions. How…
ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair
Large language model agents can repair real repository issues, but they often spend large context budgets on whole-file reads, broad searches, and long terminal outputs where useful evidence is mixed with irrelevant code…
Program RepairMIND: Lightweight and Effective Memory Injection Defense for LLM Agents via Intent-Aware Information Bottleneck
Memory-augmented LLM-based agents are vulnerable to memory injection attacks: Agents may retrieve poisoned memory from attackers, which diverts their behavior from initial user intent and finally causes task failure. How…